Audio codecs process signals inputted from one or more channels. Generally, when there is one input channel and one output channel, signals are referred to as mono signals. When there are two input channels and two output channels, signals are referred to as stereo signals. When the number of input channels and output channels are more than two, signals are called as multi-channel signals. In stereo signal coding, if signals of each channel are coded independently, then the bit-rate for transmission becomes high. But, the bit-rate can be reduced by using a stereo coding algorithm. Examples of audio coding for processing stereo signals, which will be referred to as stereo coding, include intensity stereo coding, Mid/Side (M/S) stereo coding, and parametric stereo coding.
The intensity stereo coding has been used since Moving Picture Experts Group (MPEG-1). According to psychoacoustic analysis results, stereo signals of over 2 kHz frequency are perceived not by fine structure of audio signals but by size information in a time domain. Therefore, the intensity stereo coding method transmits scale factor of right and left channel signals and sum signals of the right and left channel signals to maintain sound shape and reduce the bit rate, instead of coding and transmitting right channel signals and left channel signals, individually.
According to M/S stereo coding, the sum and subtraction of normalized right and left signals are transmitted instead of the right and left signals being transmitted. The M/S stereo coding can adjust short time delay between the right channel and the left channel, control the sound shape, and acquire a little bit of signal processing gain. The adjustable time delay is limited. However, since the time delay is longer than a time delay acoustically perceived by human beings, most of the poor sound shape problems can be resolved.
In case of parametric stereo coding, right channel signals and left channel signals are down-mixed, coded, and transmitted. To represent stereo effect, panorama, ambience, and stereo image such as time and phase difference of stereo channel are made into parameters and transmitted, too. With the parametric stereo coding, stereo signals can be represented with a small number of bits, compared to the M/S stereo coding method.
FIG. 1 shows a block diagram of a typical stereo audio coding apparatus. Referring to FIG. 1, a typical stereo coding scheme does not individually code right channel signals and left channel signals. Instead, signals of the right and left channels are down-mixed in a down-mixer 101 to be converted into mono signals. The mono signals are coded in a coder 102 and transmitted. Meanwhile, parameters are extracted in a stereo representation unit 103 to give signals a stereo effect, and transmitted.
One of the most general down-mixing methods is to sum up signals of right and left channels and divide them into two (which is (R+L)/2). For the stereo representation, scale factors are extracted and transmitted according to the intensity stereo coding method, or the difference between the two signals is coded and transmitted according to the M/S stereo coding method. According to the parametric stereo coding method, various parameters are extracted and transmitted for the stereo representation. The stereo coding has a form of a down-mixing signal coding module provided with a module for extracting stereo representation parameters.
Recently, the number of portable terminals in support of stereo input and output is increasing. The portable terminals are used to transmit not only music signals but also voice signals for conversation between users. However, the stereo effect of voice signals tends to be weaker than that of music signals. Also, since the distance between an input terminal and a speaking user is short in case of portable terminals, there is little difference between right channel signals and left channel signals during voice communication. Thus, users scarcely perceive the difference between stereo and mono. Meanwhile, in case of a portable terminal supplied with power from batteries, the battery lifecycle can be extended by reducing the amount of calculation needed for processing input signals.
Therefore, when the conventional stereo coding method described above is applied to portable terminals mainly used for transmitting/receiving voice signals, the amount of calculation needed for processing input signals increases unnecessarily. This increases power consumption and shortens battery lifecycle.